Richard Cownie (tich@pobox.com) on 8/5/10 wrote:
---------------------------
>Michael S (already5chosen@yahoo.com) on 8/5/10 wrote:
>---------------------------
>
>>Nehalem-based quad-core Xeon/2.93 without turboboost? There is no such thing. Assuming
>>you are talking about x5570, it has Max Turbo Frequency=3.333 GHz.
>>http://ark.intel.com/Product.aspx?id=37111
>>
>>As to Core2 Xeons, I am not aware of 2.93GHz parts. There are 2.83 GHz, 3.0 GHz, 3.16 GHz and 3.20 GHz.
>
>Here's /proc/cpuinfo for the two machines:
>
>model name : Intel(R) Xeon(R) CPU E7220 @ 2.93GHz
>stepping : 11
>cpu MHz : 2925.866
>cache size : 4096 KB
>

That's not a "normal" Xeon.
That's 1st-generation Core-based Xeon-MP a.k.a Tigerton castrated for lower cost and lower power by disabling half of the cores. Still, it is more expensive and good deal slower than "normal" C2D-based Xeons, esp of Penryn generation.
http://ark.intel.com/Product.aspx?id=30780
Pay attention to slow FSB.
Unless you have IBM machine, this Xeon comes with Intel 7300 chipset that is build for scalability on commercial applications rather than for good single-thread performance. In particular, "idle" memory access latency is the slowest of all Intel server/workstation/desktop products introduced since (or, may be, including) E8501.

For the application depending mostly on main memory latency E7220+7300 combo is almost worst possible representative of C2D family.
Now, I'd imagine that you were buying these chips because per socket and per core they provide the highest memory capacity. That's of course a legitimate reasons. But, IMHO, you can't extrapolate comparisons made on such relatively uncommon setup to more general C2D vs Nehalem case in which Nehalem still enjoys memory latency advantage but, esp on desktop, the difference is much much smaller.

>model name : Intel(R) Xeon(R) CPU X5570 @ 2.93GHz
>stepping : 5
>cpu MHz : 2926.152
>cache size : 8192 KB
>

That's the one I mentioned in a post above. Turboboosts to 3.33 GHz.